Research map / Detection and segmentation / CNN features
MetaFusion: Infrared and Visible Image Fusion via Meta-Feature Embedding From Object Detection
Wenda Zhao, Shigeng Xie, Fan Zhao, You He, Huchuan Lu
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace cascaded optimization with meta-feature embedding (ablation) | MetaFusion · M3FD | MI: 14.187 -> 14.511 | Improved |
| Replace cascaded optimization with meta-feature embedding (ablation) | MetaFusion · M3FD | EN: 7.089 -> 7.249 | Improved |
| Replace cascaded optimization with meta-feature embedding (ablation) | MetaFusion · M3FD | VIF: 1.345 -> 1.515 | Improved |
| Add mutual-promotion learning, increasing rounds from R = 0 to R = 2 (ablation) | MetaFusion · M3FD | MI: 14.164 -> 14.511 | Improved |
| Add mutual-promotion learning, increasing rounds from R = 0 to R = 2 (ablation) | MetaFusion · M3FD | EN: 7.078 -> 7.249 | Improved |
| Add mutual-promotion learning, increasing rounds from R = 0 to R = 2 (ablation) | MetaFusion · M3FD | VIF: 1.190 -> 1.515 | Improved |
| Add mutual-promotion learning, increasing rounds from R = 0 to R = 2 (ablation) | MetaFusion · M3FD | mAP50→95 (%): 55.6 -> 56.5 | Improved |
| Increase meta-feature embedding from one level to three levels (ablation) | MetaFusion · M3FD | VIF: 1.259 -> 1.474 | Improved |
| Use MetaFusion fusion images rather than U2Fusion images for detection training | YOLOv5s · M3FD | mAP50→95 (%): 55.7 -> 56.5 | Improved |
| Use MetaFusion rather than Tardal for fusion | Image fusion · RoadScene | MI: 15.009 -> 14.218 | Hurt |
Later papers that compared with this method
1 later accepted paper reports MetaFusion as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- E2E-MFD: Towards End-to-End Synchronous Multimodal Fusion Detection (NeurIPS 2024 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- A Deep Learning Framework for Infrared and Visible Image Fusion Without Strict Registration (IJCV 2023 · similarity 0.85)
- Target-Aware Dual Adversarial Learning and a Multi-Scenario Multi-Modality Benchmark To Fuse Infrared and Visible for Object Detection (CVPR 2022 · similarity 0.84)
- Distribution-Aligned Multimodal Fusion for Robust Object Detection (CVPR 2026 · similarity 0.83)
- P²Fusion: Prompt-based Progressive Infrared-Visible Image Fusion via Dual-Prior Distillation (ECCV 2026 · similarity 0.83)
- Event-based Visible and Infrared Fusion via Multi-task Collaboration (CVPR 2024 · similarity 0.82)
- Fine-Grained Prototypes Distillation for Few-Shot Object Detection (AAAI 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
